Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method of a system comprising a hardware processor, comprising: receiving sensor data from a sensor; responsive to a parameter of a hardware component indicated by the sensor data, trigger a multi-stage image analysis for fault determination, the multi-stage image analysis comprising: receiving first image data indicative of a computing system display output, wherein the first image data comprises a first image of a frame of the computing system display output; receiving second image data that is a scaled version of the first image data, the second image data having a lower resolution than the first image data; in a first stage of the multi-stage image analysis, comparing the second image data to reference image data corresponding to a fault display output; and performing a second stage of the multi-stage image analysis responsive to determining, based on the comparison, that a similarity of the second image data to the reference image data exceeds a similarity threshold, the second stage comprising identifying a sequence of characters in the first image data and determining whether a fault state of an operating system is present based on the identified sequence of characters.
2. The method of claim 1 , wherein the triggering of the multi-stage image analysis is responsive to a temperature indicated by the sensor data.
A system and method for multi-stage image analysis in industrial or environmental monitoring applications. The invention addresses the challenge of efficiently processing high-resolution images in real-time while conserving computational resources. The system includes a sensor network that captures image data and environmental parameters, such as temperature, from a monitored area. A processing unit performs a multi-stage image analysis, where the initial stage involves a low-resolution or simplified analysis to quickly identify regions of interest. If the analysis detects potential anomalies or significant features, a higher-resolution or more computationally intensive analysis is triggered for those regions. The triggering of this multi-stage analysis is responsive to temperature data from the sensors, allowing the system to prioritize analysis when thermal conditions may indicate higher risk or greater relevance. For example, in industrial settings, elevated temperatures might correlate with equipment malfunctions, prompting deeper image inspection. The system dynamically adjusts the analysis depth based on real-time sensor inputs, optimizing resource usage while maintaining accuracy. This approach is particularly useful in applications requiring continuous monitoring, such as manufacturing quality control, environmental surveillance, or infrastructure inspection.
3. The method of claim 1 , wherein the determining of whether the fault state of the operating system is present comprises comparing the identified sequence of characters to a reference sequence of characters in the fault display output.
This invention relates to fault detection in operating systems, specifically identifying and diagnosing fault states by analyzing display output. The problem addressed is the need for an automated and reliable way to detect operating system faults, particularly when the system is unresponsive or displays error messages. The solution involves monitoring the system's display output to identify sequences of characters that indicate a fault state. The method compares these identified sequences to predefined reference sequences known to correspond to fault conditions. If a match is found, the system determines that a fault state is present. This approach allows for early detection of system failures, enabling quicker troubleshooting and recovery. The reference sequences can be stored in a database and updated as new fault patterns are discovered. The method may also include preprocessing the display output to filter out irrelevant data, improving accuracy. By automating fault detection through display output analysis, the invention reduces manual intervention and improves system reliability. The technique is particularly useful in environments where system crashes or hangs occur frequently, such as in high-performance computing or industrial control systems. The invention ensures that faults are detected promptly, minimizing downtime and potential data loss.
4. The method of claim 1 , wherein the triggering of the multi-stage image analysis is responsive to a fan speed indicated by the sensor data.
A system and method for multi-stage image analysis in industrial environments, particularly for monitoring equipment such as fans, addresses the challenge of efficiently detecting and diagnosing faults in real-time. The system uses sensor data, including vibration, temperature, and speed measurements, to trigger a multi-stage image analysis process. The first stage involves capturing images of the equipment using one or more cameras positioned to monitor critical components. The second stage applies computer vision techniques to analyze the images for signs of wear, misalignment, or other anomalies. The third stage compares the analysis results against predefined thresholds or historical data to determine if maintenance is required. The triggering of this multi-stage analysis is specifically responsive to fan speed data, ensuring that the system prioritizes analysis when speed variations may indicate potential issues. This approach reduces unnecessary processing by focusing computational resources on conditions that are more likely to indicate faults, improving efficiency and reliability in predictive maintenance. The system may also integrate additional sensors and adjust analysis parameters dynamically based on environmental conditions or equipment load.
5. The method of claim 1 , comprising: in the first stage, determining a correlation coefficient based on the comparison of the second image data and the reference image data, wherein the determining that the similarity of the second image data to the reference image data exceeds the similarity threshold comprises determining that the correlation coefficient exceeds a correlation threshold; and wherein the identifying the sequence of characters comprises applying optical character recognition to identify the sequence of characters in the first image data.
This invention relates to image processing and optical character recognition (OCR) systems, specifically for verifying the authenticity of documents or images by comparing them to reference data. The problem addressed is the need for an automated and reliable method to detect similarities between a second image (e.g., a scanned or photographed document) and a reference image (e.g., a known authentic version) to ensure accuracy and prevent fraud. The method involves a multi-stage process. In the first stage, a correlation coefficient is calculated by comparing the second image data to the reference image data. This coefficient quantifies the similarity between the two images. If the correlation coefficient exceeds a predefined correlation threshold, it indicates that the second image is sufficiently similar to the reference image. In such cases, optical character recognition (OCR) is then applied to the first image data (which may be the same as or related to the second image) to extract and identify a sequence of characters, such as text or numbers, from the image. This extracted sequence can be used for further verification or processing. The method ensures that only images meeting a minimum similarity threshold undergo OCR, improving efficiency and accuracy in document authentication or data extraction tasks. The use of correlation coefficients provides a robust metric for similarity assessment, while OCR enables automated text extraction from verified images.
6. The method of claim 1 , further comprising: sending a fault indication of the fault state to a monitoring system.
A system and method for detecting and reporting faults in a technical infrastructure, particularly in industrial or networked environments where real-time monitoring is critical. The invention addresses the need for automated fault detection and immediate notification to ensure system reliability and minimize downtime. The method involves continuously monitoring operational parameters of a system or device to identify deviations from expected performance, which are classified as fault states. When a fault is detected, the system generates a fault indication that includes details about the nature and severity of the issue. This fault indication is then transmitted to a centralized monitoring system, which can be a supervisory control and data acquisition (SCADA) system, a network management platform, or another monitoring infrastructure. The monitoring system processes the fault indication to trigger alerts, log events, or initiate corrective actions. The method ensures timely intervention by providing real-time visibility into system health, reducing the risk of unplanned failures and improving operational efficiency. The invention is applicable in various domains, including industrial automation, telecommunications, and data center management, where proactive fault management is essential.
7. The method of claim 1 , wherein the triggering of the multi-stage image analysis is responsive to a power supply voltage indicated by the sensor data.
A system and method for multi-stage image analysis in electronic devices, particularly for optimizing power consumption and performance. The invention addresses the challenge of balancing computational efficiency with accurate image processing in resource-constrained environments, such as mobile devices or embedded systems. The method involves analyzing sensor data to determine the power supply voltage of the device. When the voltage falls below a predefined threshold, a multi-stage image analysis process is triggered. This process includes a first stage of low-power, coarse image analysis to quickly assess the image content, followed by a second stage of higher-power, detailed analysis if necessary. The system dynamically adjusts the processing stages based on the available power supply voltage, ensuring efficient use of computational resources while maintaining image processing accuracy. The method may also include additional stages of analysis, each with increasing computational complexity and power consumption, depending on the voltage level and the requirements of the image processing task. This approach allows the device to adapt its image processing capabilities in real-time, optimizing performance and battery life.
8. A system comprising: a processor; and a non-transitory storage medium storing instructions executable on the processor to: receive sensor data from a sensor; responsive to a parameter of a hardware component indicated by the sensor data, trigger a multi-stage image analysis for fault determination, the multi-stage image analysis comprising: receiving first image data indicative of a computing system display output, receiving second image data that is a scaled version of the first image data, the second image data having a lower resolution than the first image data, in a first stage of the multi-stage image analysis, comparing the second image data to reference image data corresponding to a fault display output, performing a second stage of the multi-stage image analysis responsive to determining, based on the comparison, that a similarity of the second image data to the reference image data exceeds a similarity threshold, the second stage comprising identifying a sequence of characters in the first image data and determining whether a fault state of an operating system is present based on the identified sequence of characters, and responsive to determining that the fault state is present, generate a fault indication.
The system monitors hardware components in a computing system by analyzing display outputs to detect faults. The system receives sensor data from a sensor monitoring a hardware component and, based on a parameter from the sensor data, initiates a multi-stage image analysis to determine if a fault exists. In the first stage, the system receives image data of the computing system's display output and a lower-resolution scaled version of that image data. The scaled image is compared to reference image data associated with known fault display outputs. If the similarity between the scaled image and the reference image exceeds a predefined threshold, the system proceeds to a second stage of analysis. In the second stage, the system identifies sequences of characters in the original high-resolution image data and checks whether the identified text indicates a fault state in the operating system. If a fault state is detected, the system generates a fault indication. This approach efficiently narrows down potential faults by first using low-resolution comparisons to reduce computational overhead before performing more detailed text-based analysis.
9. The system of claim 8 , wherein the instructions are executable on the processor to trigger the multi-stage image analysis responsive to any or a combination of a temperature indicated by the sensor data, a fan speed indicated by the sensor data, or a power supply voltage indicated by the sensor data.
This invention relates to a computing system with enhanced thermal management using multi-stage image analysis. The system monitors sensor data, including temperature, fan speed, and power supply voltage, to detect potential hardware issues. When abnormal conditions are detected, the system triggers a multi-stage image analysis process to further investigate the problem. The first stage involves capturing an image of the internal components, such as circuit boards or connectors, using an integrated camera. The second stage processes the image to identify defects, such as loose connections, damaged components, or excessive dust accumulation. The system then generates an alert or takes corrective action based on the analysis. The multi-stage approach ensures accurate detection while minimizing false positives. The system may also log the sensor data and analysis results for historical tracking and predictive maintenance. This solution improves hardware reliability by proactively identifying and addressing thermal and electrical issues before they cause system failures. The invention is particularly useful in high-performance computing environments where component degradation can lead to costly downtime.
10. The system of claim 8 , wherein the first image data is indicative of a single frame of a display output stream of a computing system.
A system for processing image data from a computing system's display output stream is described. The system captures a single frame of the display output stream as first image data, which represents the visual content being rendered by the computing system at a specific moment in time. This captured frame is then analyzed to extract relevant information, such as visual features, text, or other display elements. The system may further process this information for applications like monitoring, analysis, or interaction with the displayed content. The captured frame can be used independently or in conjunction with additional frames to track changes, detect events, or perform other display-related tasks. The system ensures accurate representation of the display output by precisely capturing the frame data, which may include pixel values, color information, or other visual attributes. This approach enables real-time or post-processing of the display output for various purposes, such as debugging, user interface analysis, or automated testing. The system may integrate with other components to enhance functionality, such as image recognition modules or data storage systems, to provide comprehensive display output processing capabilities.
11. The system of claim 8 , further comprising a communication interface to send the fault indication to a monitoring system.
A system for detecting and reporting faults in an industrial or electronic system monitors operational parameters to identify deviations from expected performance. The system includes sensors or measurement devices that collect data on variables such as voltage, current, temperature, or mechanical stress. A processing unit analyzes this data using predefined thresholds or machine learning models to detect anomalies indicative of faults. When a fault is detected, the system generates a fault indication, which may include details such as the type of fault, its severity, and its location within the system. The system also includes a communication interface to transmit this fault indication to a monitoring system, which may be a centralized control unit, a remote server, or a human-machine interface. The monitoring system can then take corrective actions, such as alerting operators, triggering automated repairs, or logging the fault for maintenance. This system improves fault detection accuracy and response time, reducing downtime and maintenance costs in industrial or electronic applications.
12. The system of claim 8 , wherein the instructions are executable on the processor to: receive a display output produced by the operating system; generate the first image data based on the display output.
A system for capturing and processing display output from an operating system includes a processor and memory storing instructions. The system receives a display output generated by the operating system and processes this output to generate first image data. The system may also include a display device for presenting the first image data, where the display device is configured to receive the first image data and render it for visual output. Additionally, the system may include a camera for capturing a second image of a physical environment, where the second image data is processed to detect objects or features in the environment. The system may further include a memory for storing the first and second image data, and a communication interface for transmitting the image data to a remote server. The system may also include a user interface for receiving user input to control the display or processing of the image data. The instructions executed by the processor enable the system to perform these functions, including capturing, processing, and displaying the image data in real-time or near real-time. The system may be used in applications such as augmented reality, virtual reality, or other display-based systems where real-time processing of display output is required.
13. The system of claim 8 , wherein the instructions are executable on the processor to apply optical character recognition on the first image data to identify the sequence of characters.
A system for processing image data includes a processor and memory storing instructions executable by the processor. The system captures a first image of a physical object, such as a document or label, using an imaging device. The instructions process the first image data to extract a sequence of characters, such as text or symbols, from the image. Optical character recognition (OCR) is applied to the first image data to identify and convert the sequence of characters into machine-readable text. The system may also capture a second image of the same physical object from a different perspective or under different conditions, and the instructions align the first and second images to improve character recognition accuracy. The extracted text may be used for data entry, document processing, or other applications requiring text extraction from physical objects. The system enhances automation in environments where manual data entry is error-prone or inefficient, such as logistics, inventory management, or document digitization. The OCR process may include preprocessing steps like noise reduction, contrast adjustment, or skew correction to improve recognition performance. The system may also validate the extracted text against known formats or databases to ensure accuracy.
14. The system of claim 9 , wherein the instructions are executable on the processor to trigger the multi-stage image analysis further based on an operating state of a computing system.
A system for multi-stage image analysis includes a processor and memory storing instructions that, when executed, perform image analysis in multiple stages. The system processes an input image through a first stage to generate a first output, then processes the first output through a second stage to generate a second output. The second stage may include a neural network or other machine learning model. The system may also include a user interface for displaying the second output or intermediate results. The instructions can trigger the multi-stage analysis based on the operating state of the computing system, such as available processing power, memory usage, or other system conditions. This ensures efficient resource utilization while maintaining analysis accuracy. The system may also adjust analysis parameters, such as resolution or model complexity, based on the operating state to optimize performance. The multi-stage approach allows for progressive refinement of image analysis results, balancing speed and accuracy according to system capabilities.
15. A non-transitory machine-readable storage medium comprising instructions that upon execution cause a system to: receive sensor data from a sensor; responsive to a parameter of a hardware component indicated by the sensor data, trigger a multi-stage image analysis for fault determination, the multi-stage image analysis comprising: receiving first image data indicative of a computing system display output, receiving second image data that is a scaled version of the first image data, the second image data having a lower resolution than the first image data, in a first stage of the multi-stage image analysis, comparing the second image data to reference image data corresponding to a fault display output, performing a second stage of the multi-stage image analysis responsive to determining, based on the comparison, that a similarity of the second image data to the reference image data exceeds a similarity threshold, the second stage comprising identifying a sequence of characters in the first image data and determining whether a fault state of an operating system is present based on the identified sequence of characters, and generate a fault indication responsive to determining that the fault state of the operating system is present.
This invention relates to automated fault detection in computing systems using image analysis. The system monitors sensor data from hardware components to detect potential faults. When a parameter from the sensor data indicates a possible issue, the system initiates a multi-stage image analysis process. First, it captures an image of the computing system's display output. The system then generates a lower-resolution version of this image and compares it to reference images associated with known fault conditions. If the similarity between the low-resolution image and a reference fault image exceeds a predefined threshold, the system proceeds to a second analysis stage. In this stage, the system examines the original high-resolution image to identify text or character sequences that may indicate a specific fault state in the operating system. If such a fault state is confirmed, the system generates a fault indication to alert users or trigger corrective actions. The multi-stage approach improves efficiency by first performing a quick, low-resolution comparison before conducting more detailed text analysis, reducing computational overhead while maintaining accuracy in fault detection.
16. The non-transitory machine-readable storage medium of claim 15 , wherein the instructions upon execution cause the system to: trigger the multi-stage image analysis responsive to any or a combination of a temperature indicated by the sensor data, a fan speed indicated by the sensor data, or a power supply voltage indicated by the sensor data.
This invention relates to a system for monitoring and analyzing sensor data in a computing environment, particularly for triggering multi-stage image analysis based on specific sensor conditions. The system includes a non-transitory machine-readable storage medium containing instructions that, when executed, cause a computing device to perform operations. These operations involve receiving sensor data from one or more sensors associated with the computing device, where the sensor data includes measurements such as temperature, fan speed, and power supply voltage. The system then processes this sensor data to determine whether predefined conditions are met. If any of these conditions are satisfied—such as a temperature exceeding a threshold, a fan speed falling outside an expected range, or a power supply voltage deviating from a nominal value—the system triggers a multi-stage image analysis process. This analysis may involve capturing images of the computing device's internal components, analyzing these images for anomalies, and generating alerts or taking corrective actions based on the findings. The invention aims to enhance system reliability and maintenance by proactively detecting potential hardware issues through automated sensor-based triggers.
17. The non-transitory machine-readable storage medium of claim 16 , wherein the instructions upon execution cause the system to: apply optical character recognition to identify the sequence of characters in the first image data.
This invention relates to image processing systems that extract text from images. The problem addressed is accurately identifying and extracting text sequences from image data, particularly in scenarios where the text may be distorted, low-resolution, or embedded in complex backgrounds. The invention involves a non-transitory machine-readable storage medium containing instructions that, when executed, enable a system to process image data. The system captures a first image containing text and applies optical character recognition (OCR) to identify the sequence of characters in the image. The OCR process converts the visual text into machine-readable text data, allowing for further analysis or storage. The system may also compare the extracted text with reference data to verify accuracy or detect anomalies. The invention improves upon existing OCR techniques by enhancing text recognition in challenging conditions, such as varying fonts, lighting, or noise. The storage medium may be part of a larger system that includes image capture devices, processing units, and output interfaces to display or transmit the recognized text. The invention is applicable in document digitization, automated data entry, and text extraction from images in fields like healthcare, finance, and logistics.
18. The non-transitory machine-readable storage medium of claim 16 , wherein the system comprises a processor that is separate from a processor of a computing system comprising the operating system.
A system for managing computing resources includes a non-transitory machine-readable storage medium storing instructions that, when executed by a processor, cause the processor to monitor the computing system's operating system for resource usage data. The system identifies resource-intensive processes and dynamically allocates additional resources to those processes based on predefined criteria. The processor executing these instructions is separate from the computing system's main processor, ensuring independent operation and reducing the load on the primary processing unit. This approach optimizes resource allocation, improves system performance, and prevents bottlenecks by offloading monitoring and management tasks to a dedicated processor. The system may also include mechanisms to adjust resource allocation in real-time based on changing workload demands, ensuring efficient utilization of available hardware. By separating the monitoring and management functions from the primary computing system, the solution enhances stability and responsiveness, particularly in environments with high computational demands. The storage medium may further include instructions for logging resource usage patterns and generating reports to assist in system optimization and troubleshooting. This design is particularly useful in data centers, cloud computing environments, and high-performance computing systems where efficient resource management is critical.
Unknown
December 15, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.